Leveraging SQL with User Defined Aggregation to Efficiently Merge Inverted Indexes Stored as Tables

1. A method comprising: a DBMS storing an index table and a staging table that stores changes to be made to said index table; said DBMS executing an execution plan for executing a query for updating said index table, said query specifying: to group rows from said staging table into groups according to a grouping key, each group of said groups comprising respective one or more rows; to apply an aggregate operator that is defined by a database dictionary of said DBMS to said groups; wherein executing said execution plan comprises executing said aggregate operator, wherein executing said aggregator operator includes: for each group of said groups, updating one or more rows of said index table based on changes recorded in the respective one or more rows comprising said each group.

2. The method of claim 1 , wherein said execution plan specifies multiple work granules, each work granule of said multiple work granules specifying to aggregate said groups of one or more rows; and wherein executing said execution plan includes multiple processes executing said multiple work granules in parallel.

3. The method of claim 2 wherein work granules for each group are executed by the same process.

4. The method of claim 1 , wherein executing said aggregate operator further includes committing one or more updates to said index table.

5. The method of claim 1 , wherein executing said aggregate operator further includes deleting rows from said staging table.

6. The method of claim 1 , further comprising: the DBMS storing a second staging table; and in response to executing the query, storing changes to said index table in said second staging table.

7. The method of claim 1 wherein updating one or more rows of said index table comprises appending data to said one or more rows of said index table.

8. The method of claim 7 further comprising: determining said one or more rows has insufficient space to append data; and creating a new row in said index table to store said data.

9. The method of claim 1 wherein executing said aggregate operator comprises, for each group in said groups: iterating through each row of a respective one or more rows; determining one or more changes recorded in said each row; and updating a row of said index table based on the one or more changes recorded in said each row.

10. The method of claim 1 wherein the DBMS periodically automatically executes said query.

11. The method of claim 1 wherein the index table is an inverted index and said grouping key corresponds to keywords mapped by the inverted index.

12. One or more non-transitory computer-readable media storing instructions, wherein the instructions include: instructions which, when executed by one or more hardware processors, cause a DBMS storing an index table and a staging table that stores changes to be made to said index table; instructions which, when executed by one or more hardware processors, cause said DBMS executing an execution plan for executing a query for updating said index table, said query specifying: to group rows from said staging table into groups according to a grouping key, each group of said groups comprising respective one or more rows; to apply an aggregate operator that is defined by a database dictionary of said DBMS to said groups; wherein executing said execution plan comprises executing said aggregate operator, wherein executing said aggregator operator includes: for each group of said groups, updating one or more rows of said index table based on changes recorded in the respective one or more rows comprising said each group.

13. The one or more non-transitory computer-readable media of claim 12 , wherein said execution plan specifies multiple work granules, each work granule of said multiple work granules specifying to aggregate said groups of one or more rows; and wherein executing said execution plan includes multiple processes executing said multiple work granules in parallel.

14. The one or more non-transitory computer-readable media of claim 13 wherein work granules for each group are executed by the same process.

15. The one or more non-transitory computer-readable media of claim 12 , wherein executing said aggregate operator further includes committing one or more updates to said index table.

16. The one or more non-transitory computer-readable media of claim 12 , wherein executing said aggregate operator further includes deleting rows from said staging table.

17. The one or more non-transitory computer-readable media of claim 12 , the instructions further including: instructions which, when executed by one or more hardware processors, cause the DBMS storing a second staging table; and instructions which, when executed by one or more hardware processors, cause in response to executing the query, storing changes to said index table in said second staging table.

18. The one or more non-transitory computer-readable media of claim 12 wherein updating one or more rows of said index table comprises appending data to said one or more rows of said index table.

19. The one or more non-transitory computer-readable media of claim 18 further comprising: instructions which, when executed by one or more hardware processors, cause determining said one or more rows has insufficient space to append data; and instructions which, when executed by one or more hardware processors, cause creating a new row in said index table to store said data.

20. The one or more non-transitory computer-readable media of claim 12 wherein executing said aggregate operator comprises, for each group in said groups: iterating through each row of a respective one or more rows; determining one or more changes recorded in said each row; and updating a row of said index table based on the one or more changes recorded in said each row.

21. The one or more non-transitory computer-readable media of claim 12 wherein the DBMS periodically automatically executes said query.

22. The one or more non-transitory computer-readable media of claim 12 wherein the index table is an inverted index and said grouping key corresponds to keywords mapped by the inverted index.